Involutive Yang-Baxter Group (IYB-group)

• IYB-groups are finite groups derived from involutive, non-degenerate solutions to the Yang–Baxter equation, defined via bijective 1-cocycles and brace theory.
• They admit multiple equivalent formulations including structure groups, permutation groups, and cohomological perspectives that bridge algebraic constructs.
• Their study reveals practical insights into constructing finite quotients, analyzing orderability and multipermutation properties, and resolving classification problems.

An involutive Yang–Baxter group is a group attached to a finite involutive, non-degenerate set-theoretic solution of the Yang–Baxter equation. The term is used in several closely related ways in the literature: one convention identifies an IYB-group with the structure group $G_{(X,r)}$ of a finite involutive solution; another identifies it with the permutation group $\mathcal{G}(X,r)$ generated by the left actions of such a solution; and a cohomological or brace-theoretic convention defines it as a finite group admitting a bijective $1$-cocycle, equivalently as the multiplicative group of a finite left brace (Lebed et al., 2017, Meng et al., 8 Sep 2025, David et al., 2014).

1. Terminological scope and basic objects

A set-theoretic solution of the Yang–Baxter equation is a pair $(X,r)$ with $X$ a set and

$$r\colon X\times X\to X\times X,\qquad r(x,y)=(\sigma_x(y),\tau_y(x)),$$

satisfying

$r_1r_2r_1=r_2r_1r_2$

on $X^3$, where $r_1=r\times \mathrm{Id}_X$ and $r_2=\mathrm{Id}_X\times r$. In the finite theory relevant here, one usually assumes $\mathcal{G}(X,r)$0 bijective and non-degenerate, meaning that for each $\mathcal{G}(X,r)$1 the maps $\mathcal{G}(X,r)$2 are bijections, and one often further assumes involutivity,

$\mathcal{G}(X,r)$3

For involutive solutions, strong symmetry relations connect $\mathcal{G}(X,r)$4 and $\mathcal{G}(X,r)$5, and this class is central in the theory of IYB-groups (Lebed et al., 2017).

The literature uses the following principal conventions.

Convention	Defining object	Source
Structure-group convention	$\mathcal{G}(X,r)$6 for a finite, non-degenerate, involutive solution	(Lebed et al., 2017)
Permutation-group convention	$\mathcal{G}(X,r)$7 or $\mathcal{G}(X,r)$8 for such a solution	(Meng et al., 8 Sep 2025)
Cohomological/brace convention	finite group with a bijective $\mathcal{G}(X,r)$9-cocycle; equivalently multiplicative group of a finite left brace	(David et al., 2014)

Under the structure-group convention, one defines

$1$0

Under the permutation-group convention, one defines

$1$1

or equivalently $1$2. Under the cohomological convention, a finite group $1$3 is IYB if there exists a finite abelian $1$4-module $1$5 with $1$6 and a bijective $1$7-cocycle $1$8; this is equivalent to saying that $1$9 is the multiplicative group of a finite left brace (Meng et al., 8 Sep 2025, Bachiller et al., 2013, David et al., 2014).

2. Structure groups, finite quotients, and the involutive case

For involutive solutions, the natural map $(X,r)$0 is injective, so involutive solutions are the prototype of injective solutions. In this setting, Gateva-Ivanova and Van den Bergh showed that $(X,r)$1 is a group of $(X,r)$2-type and a Bieberbach group, while Chouraqui showed that these groups are Garside. Dehornoy then attached to every finite involutive non-degenerate solution a finite Coxeter-like quotient playing for $(X,r)$3 the role that a Coxeter group plays for an Artin–Tits group (Lebed et al., 2017, Dehornoy, 2013).

A generalization valid for all finite non-degenerate solutions constructs a finite quotient

$(X,r)$4

where $(X,r)$5 is a free abelian normal subgroup of finite index generated by suitable twisted powers $(X,r)$6. If $(X,r)$7 denotes the number of orbits of the structure rack, then $(X,r)$8 has rank $(X,r)$9, the quotient $X$0 is finite, and there is a short exact sequence

$X$1

Moreover, $X$2 is injective if and only if the composite $X$3 is injective. In the involutive case the structure rack is trivial, $X$4, and $X$5 recovers, up to minor refinements, Dehornoy’s Coxeter-like group (Lebed et al., 2017).

The same framework gives an injectivization procedure for arbitrary finite solutions. Defining

$X$6

one obtains a quotient solution $X$7 that is injective and has structure group naturally isomorphic to $X$8. For involutive solutions this operation is trivial, because injectivity already holds. This places involutive solutions inside a broader theory in which injective solutions inherit many of the structural features previously associated mainly with IYB-groups (Lebed et al., 2017).

Another rigid invariant is the rank of the abelianization. If $X$9 denotes the number of orbits of the solution under the group generated by all $$r\colon X\times X\to X\times X,\qquad r(x,y)=(\sigma_x(y),\tau_y(x)),$$0 and $$r\colon X\times X\to X\times X,\qquad r(x,y)=(\sigma_x(y),\tau_y(x)),$$1, then

$$r\colon X\times X\to X\times X,\qquad r(x,y)=(\sigma_x(y),\tau_y(x)),$$2

In particular, for an indecomposable involutive solution, the abelianization of its structure group has rank $$r\colon X\times X\to X\times X,\qquad r(x,y)=(\sigma_x(y),\tau_y(x)),$$3. The torsion in $$r\colon X\times X\to X\times X,\qquad r(x,y)=(\sigma_x(y),\tau_y(x)),$$4 is not determined in general and is explicitly identified as an open problem (Lebed et al., 2017).

3. Cohomological and brace-theoretic formulations

The cohomological formulation of IYB-groups begins with groups of $$r\colon X\times X\to X\times X,\qquad r(x,y)=(\sigma_x(y),\tau_y(x)),$$5-type. A subgroup

$$r\colon X\times X\to X\times X,\qquad r(x,y)=(\sigma_x(y),\tau_y(x)),$$6

is of $$r\colon X\times X\to X\times X,\qquad r(x,y)=(\sigma_x(y),\tau_y(x)),$$7-type if the restriction of the natural projection

$$r\colon X\times X\to X\times X,\qquad r(x,y)=(\sigma_x(y),\tau_y(x)),$$8

is bijective. Passing to the kernel $$r\colon X\times X\to X\times X,\qquad r(x,y)=(\sigma_x(y),\tau_y(x)),$$9 of the action of $r_1r_2r_1=r_2r_1r_2$0 on $r_1r_2r_1=r_2r_1r_2$1, one obtains a finite quotient

$r_1r_2r_1=r_2r_1r_2$2

and a finite abelian group

$r_1r_2r_1=r_2r_1r_2$3

together with a bijective $r_1r_2r_1=r_2r_1r_2$4-cocycle

$r_1r_2r_1=r_2r_1r_2$5

A finite group $r_1r_2r_1=r_2r_1r_2$6 is an IYB-group precisely when such a triple $r_1r_2r_1=r_2r_1r_2$7 exists (David et al., 2014).

Brace theory packages the same information into two compatible group laws. A left brace is a set $r_1r_2r_1=r_2r_1r_2$8 with $r_1r_2r_1=r_2r_1r_2$9 abelian, $X^3$0 a group, and

$X^3$1

The associated action is

$X^3$2

A major result quoted in the brace literature is that a group is IYB if and only if it is the multiplicative group of a finite left brace. Conversely, every involutive non-degenerate set-theoretic solution determines such a brace structure on its associated group-theoretic objects (Meng et al., 8 Sep 2025, Bachiller et al., 2013).

There is also an augmentation-ideal characterization. A finite group $X^3$3 is IYB if and only if there exists a left ideal

$X^3$4

such that the set $X^3$5 is a system of representatives for the cosets of $X^3$6 in $X^3$7. Equivalently, the map

$X^3$8

is a bijective $X^3$9-cocycle. This viewpoint immediately implies that every IYB-group is solvable, since preimages of submodules under $r_1=r\times \mathrm{Id}_X$0 are subgroups and produce Hall subgroups of all relevant orders (Eisele, 2013).

These formulations are compatible with the permutation-group viewpoint. For a solution $r_1=r\times \mathrm{Id}_X$1, the structure brace $r_1=r\times \mathrm{Id}_X$2 maps onto the permutation group, and the quotient by the socle yields the permutation brace. In the involutive case this realizes the permutation group as a brace quotient and places both the structure-group convention and the finite-group convention within the same brace-theoretic framework (Castelli et al., 2023).

4. Multipermutation, orderability, and other constraints

A retraction of a solution is obtained by identifying elements with the same left and right action. Iterating this operation defines multipermutation level: $r_1=r\times \mathrm{Id}_X$3 is multipermutation if $r_1=r\times \mathrm{Id}_X$4 has one element for some $r_1=r\times \mathrm{Id}_X$5. For finite involutive non-degenerate solutions, this notion is reflected very sharply in the associated groups. The structure group satisfies

$r_1=r\times \mathrm{Id}_X$6

Thus within the involutive theory, multipermutation solutions are exactly those whose structure groups admit a left order (Bachiller et al., 2017).

Orderability becomes even more rigid when combined with the structure-group results for arbitrary finite solutions. Every $r_1=r\times \mathrm{Id}_X$7 is virtually abelian, and a finitely generated virtually abelian group is biorderable only when it is free abelian. Hence

$r_1=r\times \mathrm{Id}_X$8

For involutive solutions there is a further refinement: if $r_1=r\times \mathrm{Id}_X$9 is abelian, then the solution is trivial,

$r_2=\mathrm{Id}_X\times r$0

Accordingly, among IYB-groups in the structure-group sense, the only biorderable ones are the free abelian groups coming from trivial involutive solutions (Lebed et al., 2017).

Diffuseness is controlled by the same multipermutation condition. For involutive solutions,

$r_2=\mathrm{Id}_X\times r$1

Since diffuse groups are locally indicable in the amenable virtually abelian setting, non-multipermutation involutive solutions produce structure groups that are neither diffuse nor left-orderable. This sharply separates “retractable” and “irretractable” behavior at the group level (Lebed et al., 2017).

On the permutation-group side, abelian IYB-groups force retractability. Finite involutive solutions with abelian associated IYB-group are retractable, and for each positive integer $r_2=\mathrm{Id}_X\times r$2 there exists a finite square-free multipermutation solution of level $r_2=\mathrm{Id}_X\times r$3 whose associated IYB-group is an elementary abelian $r_2=\mathrm{Id}_X\times r$4-group. This answers a problem of Gateva-Ivanova and Cameron and shows that very high multipermutation level is compatible with very small commutator structure in the permutation group (Cedo et al., 2012).

A different restriction concerns permutation actions. The permutation group of a finite non-degenerate involutive solution never acts as a Frobenius group on the underlying set. One consequence is that if an indecomposable solution has dihedral permutation group $r_2=\mathrm{Id}_X\times r$5 with odd $r_2=\mathrm{Id}_X\times r$6, then the underlying set has cardinality $r_2=\mathrm{Id}_X\times r$7. For even $r_2=\mathrm{Id}_X\times r$8, the paper exhibits counterexamples to the analogous statement, including a $r_2=\mathrm{Id}_X\times r$9-element indecomposable solution with permutation group $\mathcal{G}(X,r)$00 (Kanrar et al., 2023).

5. Constructions, classification, and simple solutions

One source of new IYB-groups comes from powering constructions on solutions. From a fixed involutive non-degenerate solution $\mathcal{G}(X,r)$01, one can construct solutions $\mathcal{G}(X,r)$02 on Cartesian powers $\mathcal{G}(X,r)$03. Their permutation groups satisfy

$\mathcal{G}(X,r)$04

and under mild hypotheses—such as the existence of $\mathcal{G}(X,r)$05 with $\mathcal{G}(X,r)$06, or $\mathcal{G}(X,r)$07 when $\mathcal{G}(X,r)$08 is finite—this subgroup is the whole original IYB-group. Thus a single IYB-group can often be realized by infinitely many larger solutions (Bachiller et al., 2013).

The permutation-brace viewpoint yields finer structure theorems. A variant $\mathcal{G}(X,r)$09 of the multipermutation level coincides with the multipermutation level of the permutation skew brace, contrary to the usual one-step inequality for $\mathcal{G}(X,r)$10. The same approach gives a description of all finite indecomposable involutive solutions with abelian permutation group, and for multipermutation level $\mathcal{G}(X,r)$11 it yields the precise number of isomorphism classes of such solutions of a given size. The classification is expressed in terms of quotients of one-generated two-sided braces and orbit formulas on explicit matrix sets (Castelli et al., 2023).

Simple solutions admit a brace-theoretic characterization. A finite simple involutive solution is either the unique indecomposable solution of prime cardinality or, if its size is not prime, it is irretractable and indecomposable, and its associated left brace $\mathcal{G}(X,r)$12 has a unique minimal ideal acting transitively on a transitive cycle base. Equivalently, every non-trivial ideal of the associated brace acts transitively on that base. This gives a criterion for recognizing which IYB-groups arise from finite simple solutions (Castelli, 2022).

Recent constructions extend this picture. A new class of indecomposable, irretractable, involutive, non-degenerate solutions has been built from data $\mathcal{G}(X,r)$13, yielding solutions on $\mathcal{G}(X,r)$14 with necessary and sufficient conditions for simplicity. For a rich subclass, the permutation groups are determined as left braces, and in the finite case these solutions have square cardinality. A second construction in the same paper gives finite simple solutions of non-square cardinality whose permutation groups are simple left braces (Cedo et al., 2024).

Another influential family consists of irretractable square-free solutions that are strong twisted unions of multipermutation solutions of level at most $\mathcal{G}(X,r)$15. Their natural left brace on the permutation group has trivial socle, and for finite members of this family the structure groups are not poly-$\mathcal{G}(X,r)$16. This family contains Vendramin’s counterexample to Gateva-Ivanova’s Strong Conjecture and supplies many further counterexamples (Bachiller et al., 2015).

6. Finite-group existence theorems and the present frontier

From the finite-group standpoint, one line of work studies closure properties of the IYB condition. If $\mathcal{G}(X,r)$17 is nilpotent of class two and $\mathcal{G}(X,r)$18 is an IYB-group of order coprime to $\mathcal{G}(X,r)$19, then $\mathcal{G}(X,r)$20 is IYB; equivalently, $\mathcal{G}(X,r)$21 is IYB if and only if $\mathcal{G}(X,r)$22 is IYB. This gives a broad supply of examples and implies, in particular, that Hertweck’s counterexample to the isomorphism problem for $\mathcal{G}(X,r)$23 and all of its subgroups of the same form are IYB. The same paper gives an explicit equivariant IYB-structure on a specific class-two $\mathcal{G}(X,r)$24-group $\mathcal{G}(X,r)$25 and proves its uniqueness up to isomorphism (Eisele, 2013).

The broad classification problem has evolved over time. Earlier work treated as open whether every finite solvable group is IYB, whereas the later cohomological synthesis notes that Bachiller showed not every finite solvable group is IYB by constructing a finite nilpotent group with that property. In that setting, the finite IYB condition is encoded by a bijective $\mathcal{G}(X,r)$26-cocycle into a finite module, and groups of $\mathcal{G}(X,r)$27-type appear as infinite coverings of finite IYB-groups. The paper develops a lifting criterion for $\mathcal{G}(X,r)$28-cocycles and recovers substantial families of IYB-groups, including finite nilpotent groups of class $\mathcal{G}(X,r)$29, finite abelian-by-cyclic groups, and solvable groups of $\mathcal{G}(X,r)$30-type (David et al., 2014).

The most recent group-theoretic advance in the data concerns finite soluble groups whose Sylow subgroups have nilpotency class at most two. If such a group $\mathcal{G}(X,r)$31 has nilpotent residual $\mathcal{G}(X,r)$32 that is $\mathcal{G}(X,r)$33-free, then $\mathcal{G}(X,r)$34 is an IYB-group. There is also a complementary theorem: if all Sylow $\mathcal{G}(X,r)$35-subgroups of $\mathcal{G}(X,r)$36 are isomorphic to $\mathcal{G}(X,r)$37, then $\mathcal{G}(X,r)$38 is again an IYB-group. The proofs rely on an $\mathcal{G}(X,r)$39-decomposition into nilpotent factors, equivariant IYB-structures on the factors, and a glueing theorem assembling these data into a global brace structure. This answers Cedó–Okniński’s question positively for a large subclass of finite soluble groups with Sylow subgroups of class at most two (Meng et al., 8 Sep 2025).

Taken together, these developments show that “IYB-group” designates not a single isolated construction but a network of equivalent or adjacent formalisms—structure groups, permutation groups, brace multiplicative groups, and bijective-cocycle groups—organized around finite involutive non-degenerate solutions of the Yang–Baxter equation. The current theory combines explicit constructions, quotient and retraction techniques, orderability and diffuseness criteria, brace-theoretic ideal structure, and finite-group existence theorems into a coherent algebraic framework (Lebed et al., 2017).